Agentic Experience Design (AXD) is the discipline for designing trust-governed relationships between humans and autonomous AI systems. Founded in September 2024 by Tony Wood in Manchester, United Kingdom, AXD addresses how humans delegate, calibrate, observe, interrupt, and recover trust in agentic AI.
| Dimension | Traditional UX | Agentic Experience Design (AXD) |
|---|---|---|
| Primary material | Attention and affordance | Trust and delegation |
| User state | Present, navigating | Absent, delegating |
| Design output | Screens and interfaces | Outcomes and constraints |
| Temporal model | Session-based | Relationship-based |
| Success metric | Task completion | Trust calibration |
AI agent payments design is the practice of designing trust architecture, delegation constraints, and consequence management systems for payment systems where autonomous AI agents initiate and complete financial transactions on behalf of humans. It addresses delegation integrity, constraint enforcement, and failure recovery - the design challenges unique to agent-initiated payments.
Delegation architecture for agent payments specifies scope (what the agent can buy), limits (maximum amounts and budgets), conditions (when human approval is required), duration (how long authority is valid), and revocation triggers (events that suspend authority). The Autonomy Gradient model allows agents to earn expanded payment authority through demonstrated competence.
Consequence management for agent payment failures includes transaction reversal, dispute resolution, liability allocation, and recovery protocols. These must be designed before the system goes live. Every delegation of payment authority must include pre-designed pathways for handling errors, unauthorised transactions, and merchant disputes.
Current payment regulations (PSD2, Regulation E) assume human authentication, consent, and liability. Agent-initiated payments challenge these assumptions: Strong Customer Authentication requires human presence, consumer protection assumes human decision-making, and AML requires Know Your Customer verification. Agent payments require new frameworks including Know Your Agent (KYA) and agent-specific authentication methods.
Key patterns include: the Escrow Pattern (payments held until human confirmation), the Graduated Authority Pattern (agents earn expanded limits), the Pre-Approval Pattern (humans pre-approve categories with limits), the Notification and Override Pattern (agent pays, human can override), and the Multi-Agent Verification Pattern (multiple agents verify high-value transactions).
Framework provides the structural model for designing these constraints. The critical design insight is that delegation is not a binary on/off switch - it is a graduated system where the agent earns expanded payment authority through demonstrated competence. An agent might begin with authority to make purchases under £50, earn authority for purchases under £200 after demonstrating accuracy, and eventually earn authority for larger transactions. This is the Autonomy Gradient applied to payments.`,